DOI 10.1140/epje/i2008-10387-2
Regular Article
P
HYSICAL
J
OURNAL
E
Temperature-sensitive poly(N-Isopropyl-Acrylamide) microgel
particles: A light scattering study
M. Reufer1,2, P. D´ıaz-Leyva1, I. Lynch3, and F. Scheffold1,a
1 Department of Physics and Fribourg Center for Nanomaterials, University of Fribourg, CH-1700 Fribourg, Switzerland 2 Adolphe Merkle Institute, University of Fribourg, CH-1700 Fribourg, Switzerland
3 School of Chemistry and Chemical Biology, University College Dublin, Belfield, Dublin 4, Ireland
Received 11 August 2008
Published online: 25 November 2008 – c EDP Sciences / Societ`a Italiana di Fisica / Springer-Verlag 2008 Abstract. We present a light scattering study of aqueous suspensions of microgel particles consisting of poly(N-Isopropyl-Acrylamide) cross-linked gels. The solvent quality for the particles depends on temper-ature and thus allows tuning of the particle size. The particle synthesis parameters are chosen such that the resulting high surface charge of the particles prevents aggregation even in the maximally collapsed state. We present results on static and dynamic light scattering (SLS/DLS) for a highly diluted sample and for diffuse optical transmission on a more concentrated system. In the maximally collapsed state the scattering properties are well described by Mie theory for homogenous hard spheres. Upon swelling we find that a radially inhomogeneous density profile develops.
PACS. 82.70.Dd Colloids – 83.80.Kn Physical gels and microgels – 82.70.Gg Gels and sols
1 Introduction
Colloidal particles with adjustable interaction potential have been of scientific and technological interest in recent years due to their potential use to control bulk properties such as viscous flow, optical and also magnetic proper-ties [1–7]. Thermo-sensitive microgels have been widely used as model systems [2, 8–10]. These materials have also received attention due to their potential applications in drug delivery or as sensors as a result of their “re-sponsive” characteristics following changes in their envi-ronment [11]. One of the most widely studied systems is poly(N-Isopropyl-Acrylamide) (PNIPAM), a polymer which has a critical solution temperature of approximately 33◦C. PNIPAM colloids can be prepared by cross-linking PNIPAM resulting in microgel particles with tunable soft-ness (and swelling degree), these depending on the cross-link density. Another approach is to coat well-defined solid core particles with PNIPAM, thereby exploiting properties of both materials [8].
Both pure PNIPAM and PNIPAM-coated particles display properties due to the tunable network combined with properties of classical colloids, e.g. crystallization or aggregation. This is very useful to tailor colloidal systems that can be kept close to the liquid-solid transition, thus having the possibility to “temper” these materials [12]. Tempering is not possible using most “classical” colloidal
a e-mail: Frank.Scheffold@unifr.ch
systems that require a change in composition in order to cross a phase boundary. In this respect temperature-sensitive particles are ideal candidates to provide quantita-tive information about the ergodic–nonergodic-transition in systems with repulsive interactions, both for the glass transition [13–17] as well as for crystallization [18].
At temperatures well above 35◦C the PNIPAM parti-cles are collapsed and behave as solid spheres. The effec-tive interaction potential can be either attraceffec-tive due to van der Waals forces which lead to aggregation and phase separation [19], or, in the presence of surface charges, the particles can be stabilized in suspension. Upon lowering the temperature the particles swell by a factor of two to five in size [2, 20]. If the initial particle density is suffi-ciently high, the particle volume increase can drive the system from a liquid to a solid state. A number of rheo-logical studies were able to study in detail the apparent divergence of the viscosity at the transition point and the emergence of an elastic shear modulus [2, 21]. More re-cently, studies on the internal dynamics and the frequency-and shear-rate-dependent rheology close to frequency-and above the liquid-solid transition have been reported [22, 23].
In this article we discuss the temperature-dependent properties of highly cross-linked microgel particles. The focus of the present work is on a comprehensive op-tical characterization of PNIPAM microgel suspensions. We present results on static and dynamic light scatter-ing (SLS/DLS) for a highly diluted sample and for dif-fuse optical transmission on a more concentrated system.
A detailed understanding of the temperature-dependent structural properties of individual particles is a key in-gredient for the study of the dynamic properties and the phase behavior of dense suspensions. The parameters of the synthesis were chosen such that the effective surface charge prevents aggregation even in the maximally col-lapsed state. The advantage of such a charge-stabilized system is that the equilibrium (or quasi-equilibrium) prop-erties can be studied over the whole range of temperatures even at high volume fraction. In the current article we dis-cuss the structural aspects of our PNIPAM suspension. A detailed study of the high-density phase behaviour and rheological properties will be presented elsewhere [22].
The outline of this article is as follows: In Section 2 we describe the particle chemistry and the particle character-ization by scanning electron microscopy. In Section 3 we analyze the temperature dependence of the hydrodynamic radius and in Section 4 we discuss results from static light scattering on a dilute suspension. In Section 5 we study the optical properties (diffuse transmission) of a concen-trated suspension. Finally, in Section 6 the experimental results are summarized.
2 Sample preparation
We use free radical cross-linking polymerization of the monomer N-Isopropyl-Acrylamide (NIPAM) from Acros Organics (Acros Organics BVBA, Geel, Belgium) with the tetra-functional cross-linker N,N -Methylene-Bis-Acrylamide (BIS) from Fluka (Fluka Chemie GmbH, Buchs, Switzerland). The BIS molecules are essentially two acrylamide monomer units bridged by a covalent bond [24]. The polymerization reaction is initiated using the ionic salt Potassium Persulfate (KPS, Merck KGaA, Darmstadt, Germany). All chemicals are reagent grade and used without further purification, except NIPAM which is recrystallized from N-Hexane solution. The syn-thesis is performed by dissolving a mixture of monomer NIPAM and cross-linker BIS (at 21.43 mg/ml) in 145 ml deionized and filtered water1. The cross-linking ratio, de-fined by fbis ≡ [BIS]/([NIPAM] + [BIS]) is 6.7%. Such a relatively high ratio is expected to produce a rigid gel with approximately one molecule of cross-linker per 19 molecules of monomer [20, 25, 26]. The mixture (previously degassed for ∼ 30 minutes) is heated up to 80◦C un-der pure nitrogen atmosphere. Then, 72.8 mg of KPS dis-solved in 5 ml of degassed water is added to the mixture to start the polymerization reaction. The reaction proceeds for at least 4 hours at constant temperature. Finally the dilute suspension is extensively dialyzed for several days against deionized water to eliminate unreacted monomer excess [8, 27, 28]. The final step is carried out under nor-mal atmospheric conditions which leads to a finite solvent
1 The original batch further contained a small quantity of
TiO2 nanoparticles in an initial attempt to create core shell particles. From transmission electron microscopy we find no trace of the TiO2 in sample after concentration and purifica-tion. Therefore the system under study consists exclusively of PNIPAM particles suspended in water.
5 mm
Fig. 1. Scanning electron microscopy picture of PNIPAM par-ticles in the collapsed state (particle radius 260± 5 nm).
ionic strength due to spontaneous dissolution of carbon dioxide in water. Dense suspensions are prepared using a rotary evaporator. This procedure increases both the par-ticle density as well as the ionic strength of the solvent. In turn, at high temperatures, we expect to obtain a charge-stabilized suspension with a strongly screened Coulomb interaction potential. We do not observe any aggregation at high temperatures which indicates that the suspension is stable.
Figure 1 shows a scanning electron micrograph of our PNIPAM particles on a solid substrate. The particles are ordered in hexagonal arrays, giving us a qualitative idea about the rather low polydispersity in the collapsed state. From the analysis of several dozen particle positions, we obtain a SEM radius of 260± 5 nm.
It is well known that even in the maximally collapsed state PNIPAM microgel particles contain a non-negligible amount of water molecules. The presence of this bound water has to be taken into account for the characterization of PNIPAM colloidal suspensions. As shown by Lele et al. [29] the bound water content of a collapsed PNIPAM gel is approximately 0.38–0.4 gram per gram of polymer which corresponds to a bound water volume fraction of approximately ΦWater= 30% if we assume a water density
of 1 g/cm3 and a bulk density of PNIPAM approximately 1.1 g/cm3 [2, 30, 31].
As suggested by Erbe et al. [32] the bulk refrac-tive index of PNIPAM for visible light can be esti-mated by extrapolating literature dn/dc values [33]. One finds n = 1.52 ± 0.01. We note that similar chem-ical structures such as Poly(N-butylmethacrylamide), Poly(N-2-methoxyethyl) methacrylamide, Poly(N-meth-ylacrylamide) all have bulk indices for visible light in the range 1.51–1.54 [34]. If we assume a value of 1.33 for the refractive index of bound water, we can estimate the re-fractive index of the particles in the maximally collapsed state based on the Maxwell-Garnett mixing rule [35, 36]. For the case of small refractive index variations it can be written as: n ΦWater· 1.33 + ΦPolymer· 1.52 = 1.46.
Fig. 2. Size characterization by dynamic light scattering (DLS). Full squares: temperature-dependent hydrodynamic ra-diusRH(error bars denote spread of data points for measure-ments at three different anglesθ = 60◦, 40◦, 20◦) atΦ ∼ 10−5. Full circles: mean radius of mass distributionR. Open circles: mean radiusR plus diffuse layer 2σ.
3 Hydrodynamic radius from dynamic light
scattering
We determine the hydrodynamic radius of the particles us-ing standard dynamic light scatterus-ing in the sus-ingle scatter-ing regime with λ = 532 nm (goniometer system ALV/SP-125, ALV, Germany). A low volume fraction (Φ ∼ 10−5) the colloidal suspension is filled into a cylindrical glass tube of 10 mm inner diameter. The vial is placed in the center of a cylindrical vat filled with an index-matching fluid, cis-trans decahydronaphthalene (decalin), in order to reduce stray light reflections and scattering from the vial surface. The index-matching fluid is temperature con-trolled (±0.1◦C).
From the measured time-averaged intensity correlation function, g(2)(q, τ ) the translational free diffusion is deter-mined and the particle hydrodynamic radius extracted by common procedures [37, 38]. We restrict the analysis to relatively low scattering angles (θ = 60◦, 40◦, 20◦). The experimental data, Figure 2, shows the expected strong temperature dependence. Since PNIPAM is not soluble in hot water (T > 33◦) the cross-linked PNIPAM gels col-lapse at elevated temperatures, whereas at room tempera-ture and below water is a good solvent leading to a strong swelling. As a consequence of the relatively high cross-link density, the swelling is less pronounced compared to some of the previous studies where swelling ratios up to four or five have been reported [20, 25, 26].
4 Static light scattering
We determine the particle form factor over a tempera-ture range 15◦C ≤ T ≤ 40◦C using a home-built 3D Light Scattering setup [39]. We use a solid-state laser TUI
Optics DL 100R (TUI Optics GmbH, Munich, Germany) λ = 680.4 nm, two avalanche photomultipliers Perkin Elmer SPCM-AQR-13-FC (Perkin Elmer Inc, Fremont CA, U.S.A.) and a digital multi-tau correlator (Corre-lator.com, Bridgewater NJ, U.S.A.). The sample is con-tained in a cylindrical optical glass tube of 5 mm inner di-ameter, placed in the center of a vat filled with cis/trans decalin to reduce stray light contributions.
The q dependence of the scattering intensity I(q) ∝ P (q) of a dilute sample (Φ ≈ 4×10−5) is displayed in
Fig-ure 3 for different temperatFig-ures. For uncorrelated spher-ical particles I(q) is given by the scattering form factor P (q) = k2
0 dσdΩsc, where k0 = 2πns/λ [40]. In the
high-temperature limit the experimental data can be described by a Mie calculation for a homogenous sphere of diameter 242 nm, refractive index n = 1.46 and a polydispersity of 9% [41].
The SLS particle size in the maximally collapsed state R = 242 nm is found in rather good agreement with the hydrodynamic radius RH 260 nm. The result is also con-sistent with the SEM analysis although, if under vacuum conditions all the water is released, we would actually ex-pect to find a radius of about 215 nm. There are a number of possible reasons for this slight difference. For example we cannot exclude the possibility that some residual wa-ter remains bound to the polymer or that the particles are slightly deformed leading to a flattened shape [20]. Small errors in the alignment of the electron microscope or the calibration could also explain this difference. More exper-iments would be needed to provide a conclusive answer, which is beyond the scope of this work.
Lowering the temperature the first minimum is ex-pected to shift to lower q values due to the increasing particle size [42, 26]. Such a variation is indeed observed. A quantitative description of the data, however, requires modeling of the density profile in the particle.
Stieger and coworkers [26] suggested a convolution of the density profile for a particle of radius R with a Gaus-sian of width σ, implying an effective steric radius of order RH≈ R + 2σ. An alternative approach suggested by Ma-son and Lin is the “uniform core linear shell” [43]. Both models provide an excellent fit to the data for temper-atures of T = 25◦C and below where the particles are swollen and the effective refractive index is sufficiently low for the Rayleigh-Gans-Debye (RGD) approximation (|n/nsolvent− 1| 1) to be valid. The “uniform core lin-ear shell” model, however, results in an outer cut-off radius substantially smaller than the hydrodynamic radius (data not shown). For this reason we have chosen to apply the Gaussian approximation. We think that the latter model captures better the slow and gradual decrease of the par-ticle density in the parpar-ticle corona. One should keep in mind however, as pointed out by Mason and Lin, that the missing cut-off makes this model unphysical at very large radial distances.
Both the “uniform core linear shell” and the Gaussian approximation were originally applied to neutron scatter-ing data. In the RGD limit light and small-angle neu-tron scattering on a two-component system are sensitive
Fig. 3. Temperature-dependent form factor P (q) measured under dilute conditions (Φ ≈ 4 × 10−5). Dash-dotted lines: Mie
calculations for homogenous spheres, polydispersity 9%: (RMie = 242 nm, n = 1.46, T = 40◦C), (RMie = 275 nm, n = 1.42, T = 30◦C) and (R
Mie= 350 nm,n = 1.37, T = 10◦C). Solid lines: RGD form factor for a mass distribution with mean radius
R convoluted with a Gaussian with standard deviation σ, polydispersity 9.5 ± 0.5%.
Table 1. Hydrodynamic radius RH and static light scatter-ing parameter from a best fit to a RGD form factor assumscatter-ing a gradually decreasing density profile with a mean radius R convoluted with a Gaussian of widthσ.
T (◦C) R H R σ 10 468 335 61 15 433 328 57 20 433 321 49 25 416 308 43 30 383 290 5 35 299 262 0 40 271 257 0 45 267 50 258
to the same physical properties, namely density fluctua-tions. Light scattering probes fluctuations of the refractive index (n/nsolvent− 1), whereas neutron scattering probes fluctuations in the nuclear scattering length density. Both quantities are proportional to the local polymer density. Therefore, up to a prefactor, the model by Stieger and coworkers [26] can be applied to both light and neutron scattering data. In the RGD approximation the scattering form factor is given by P (q)∝ [A(q)]2, with
A(q) = 3 exp[−(σq)2/2][sin(qR) − qR cos(qR)]/(qR)3.
(1) It is straightforward to account for polydispersity using standard procedures [38] (polydispersity values for best fit to the data are 9.5± 0.5%). We obtain an excellent fit to the data for temperatures up to T = 30◦C. T = 30◦C
is, however, a borderline case. Here, we obtain a perfect RGD fit with σ ≈ 0, while a Mie calculation with the same parameters displays differences. This inconsistency shows that the RGD condition (|n/nsolvent− 1| 1) is
not fully met. As a consequence, in this transition regime, the thickness of the diffuse corona σ is underestimated.
Figure 2 and Table 1 summarize the results of our anal-ysis. For temperatures T = 25◦C and below the results from static (open circles) and dynamic light scattering (solid squares) match almost perfectly. However in the transition regime at T = 30◦C we observe a noticeable difference.
It is worthwhile to add that our study represents one of the first attempts to quantitatively model the full form factor (including the first minimum) of PNIPAM particles using static light scattering. Our results clearly show that such data can be of similar or even better quality than neutron scattering data but with significantly less sample preparation requirements and at a fraction of the cost. This should prove very useful for future studies of these very interesting systems in particular for particle radii of about 200 nm and above. A detailed understanding of the rheological properties in the solid state for example will require quantitative input concerning the internal polymer density distribution.
5 Diffusing transmission
The optical transmission through a cell of thickness L is determined by the scattering cross-section of the in-dividual particles, the particle number density and inter-particle correlations characterized by the structure factor S(q). For dilute non-interacting systems (S(q) ≡ 1) the
optical properties can be characterized by the scattering mean free path [44]
l = 1 ρ2 π k2 0 2k0 0 P (q) q dq = 1 ρσsc, (2)
which is found to be inversely proportional to the total scattering cross-section of a particle σsc. Initially this leads
to an exponential decay of the unscattered intensity trans-mitted through a slab of thickness L: T0 = exp (−L/l). With increasing density multiple scattering contributions become important until in the case of a highly turbid sys-tem light propagates diffusively and the total transmission coefficient can be written as [44]
T ∝ l∗, L l∗. (3)
The typical length characterizing the diffusion process is the transport mean free path l∗ which can be writ-ten quite generally in terms of the mean-square scattering vector [45] l∗ l = 2 k2 0 q2 , (4)
denotes the angular average over all scattering angles, weighted by the scattering probability.
Positional correlations between particles affect the op-tical density of concentrated colloidal suspensions because they change the angular distribution of scattered light emerging from each scattering event and therefore change the value of l∗. For correlated systems we replace P (q) by the full scattering function P (q) S(q), thus we obtain
q2 = 1 k2 0 2k0 0 P (q) S(q) q3dq 2k0 0 P (q) S(q) q dq . (5)
Therefore the general expression for l∗in a correlated sus-pensions of spherical particles reads [45, 46]
l∗= k6 0 π ρ 2k0 0 P (q) S(q) q3dq −1 . (6)
We now want compare these predictions to the scatter-ing properties of our thermosenstive PNIPAM particles. A dense suspension is prepared using a rotary evaporator from the initial as-synthesized stock solution. From dry-ing and weighdry-ing we find a mass density of≈ 14.6% w/w. Under the assumption that the dry sample is absolutely water free, this would imply a volume fraction in the col-lapsed state of Φ≈ 19%. We perform a diffuse transmis-sion experiment with a frequency-doubled Nd:YV04
solid-state laser Coherent “Verdi” (Coherent Inc., Santa Clara CA, U.S.A.) operating at λ = 532 nm. The laser beam is slightly expanded to illuminate the sample with a spot size diameter of approximately 5 mm. The sample is kept in a regular glass cell with inner dimensions 10× 2 mm (Hellma, Germany). The scattered light leaving the sam-ple is collected by a single-mode optical fiber placed in transmission geometry. The transport mean free path l∗
Fig. 4. Transport mean free path l∗ as a function of
temper-ature for a concentrated system (mass density approximately 14.6% w/w). Solid line: Mie result, l∗= 56μm, for homogenous hard spheres with radius R = 242 nm at an effective volume fraction ofΦ ≈ 19%.
at each temperature is obtained by measuring the ratio of the intensity of the transmitted light through the sample compared to the value obtained for a reference sample of known value l∗ [47, 48].
Figure 4 shows the corresponding increase of the trans-port mean free path upon decreasing the temperature. The optical transport mean free path increases from l∗ = 53 μm in the high-temperature limit (T = 40◦) to
l∗= 289 μm at T = 10◦.
By lowering its density, the cross-linked gel changes its optical properties. In fact, the swollen gel has a lower optical contrast compared to the collapsed gel. Further scattering contrast is lost as particles start to touch and deform. For perfectly homogeneous particles at an effec-tive volume fraction of 1 there should be no scattering. Residual scattering can be linked to density fluctuations that are due to: a) the imperfect space filling of the PNI-PAM microgel particles; b) an inhomogeneous radial den-sity distribution of the individual particles; c) positional correlations. In our case positional correlations contribute only weakly to the increase of l∗ since the structure fac-tor peak is located at small values of q compared to the cut-off value 2· k0.
At high temperatures the system is fluid, particles are homogenous and moreover the density is only moderately high. Under these conditions it should be possible to make a quantitative prediction of l∗ using equation (6) [46, 49]. For our analysis we use the radius obtained from static light scattering, R = 242 nm. Structural correlations are taken into account using the Perkus-Yevick structure fac-tor for a monodisperse suspension of hard spheres. For pure water as a solvent (ns = 1.33) and a particle index ns= 1.46 we find a value of l∗= 56 μm (l∗= 48 μm with S(q) ≡ 1) at a suspension volume fraction of Φ = 19%. Such an exceptionally good agreement with the experi-mental result l∗ = 53 μm must be somewhat fortuitous
given the limited accuracy of our input parameters such as particle refractive index and the effective volume fraction. Finally we would like to add a comment concerning the background medium refractive index. Previous studies on mono- and biomodal suspension of polystyrene spheres (n 1.6) up to Φ 40% found good agreement between theory (Eq. (6)) and experiment using the refractive index of pure water [46, 50]. For high-refractive-index particles (n > 2) at high volume fraction (in air) it is, however, necessary to take into account an increased effective re-fractive as shown by G´omez Rivas et al. [51] and more recently Reufer and co-workers [52]. Following the latter approach, in our case we obtain nsolvent, eff 1.346 and thus l∗= 75 μm.
6 Summary and conclusions
In summary we have presented a comprehensive study of the scattering properties of a temperature tunable col-loidal system. It would be interesting to further extend our approach by using additional techniques such as Ultra Small Angle X-Ray or Neutron Scattering to access infor-mation on smaller length scales as done in some of the previous work on similar systems. The particular system presented in this work displays some interesting properties which make it very suitable to study the phase behav-ior of colloids with repulsive interactions. The particles are charge stabilized in the maximally collapsed state and therefore fully stable even at high temperatures. The op-tical transport mean free path at T = 40◦C furthermore indicates that at high temperatures the system essentially behaves as a suspension of hard spheres. This in turn will allow the system to reversibly cross the phase boundary from a dense colloidal suspension to an arrested system upon changing the temperature.
Work supported by the TOP NANO 21 Initiative, Grant No. 5971.2, the Swiss National Science Foundation, projects No. 200020-117762 and No. 200020-117755, and the Marie Curie network Grant No. MRTN-CT2003-504712. Authors also thank Christoph Neururer and Daniela Curdy for SEM fa-cility and synthesis assistance, Peter Schurtenberger, James Harden, Reinhard Sigel, Nasser Ben Braham, Veronique Trappe, Joaquim Clara Rahola, Frederic Cardinaux and Pavel Zakharov for discussions.
References
1. J. Wu, Z. Hu, in Encyclopedia of Nanoscience and Nan-otechnology edited by J.A. Schwartz, C.I. Contescu, K. Putyera (Marcel Dekker Inc., 2004), pp. 1967-1976. 2. H. Senff, W. Richtering, J. Chem. Phys. 111, 1705 (1999). 3. C. Wu, S. Zhou, Macromolecules 30, 574 (1997).
4. J.D. Debord, S. Eustis, S.B. Debord et al., Adv. Mater. 14, 658 (2002).
5. L.F. Rojas-Ochoa, J.M. Mendez-Alcaraz, J.J. S´aenz, P. Schurtenberger, F. Scheffold, Phys. Rev. Lett. 93, 073903 (2004).
6. A.M. Schmidt, Colloid Polym. Sci. 285, 953 (2007). 7. M. Karg, I. Pastoriza-Santos, J. Perez-Juste, T. Hellweg,
L.M. Liz-Marzan, Small 3, 1222 (2007).
8. N. Dingenouts, Ch. Norhausen, M. Ballauff, Macro-molecules 31, 8912 (1998).
9. K. Otake, H. Inomata, M. Konno, S. Saito, Macro-molecules 23, 283 (1990).
10. X. Wu, R.H. Pelton, A.E. Hamielec et al., Colloid Polym. Sci. 272, 467 (1994).
11. M. Shibayama, Macromol. Chem. Phys. 199, 1 (1998). 12. T. Hellweg, C.D. Dewhurst, E. Bruckner, K. Kratz, W.
Eimer, Colloid Polym. Sci. 278, 972 (2000).
13. T. Eckert, E. Bartsch, Phys. Rev. Lett. 89, 125701 (2002). 14. P.N. Pusey, W. van Megen, Nature 320, 340 (1986). 15. Ch. Beck, W. H¨artl, R. Hempelmann, J. Chem. Phys. 111,
8209 (1999).
16. J.J. Crassous, M. Siebenburger, M. Ballauff et al., J. Chem. Phys. 125, 204906 (2006).
17. A.N.S. John, V. Breedveld, L.A. Lyon, J. Phys. Chem. B 111, 796 (2007).
18. L.A. Lyon, J.D. Debord, S.B. Debord et al., J. Phys. Chem. B 108, 19099 (2004).
19. J.Z. Wu, B. Zhou, Z.B. Hu, Phys. Rev. Lett. 90, 048304 (2003).
20. K. Kratz, T. Hellweg, W. Eimer, Polymer 42, 6631 (2000). 21. I. Deike, M. Ballauff, N. Willenbacher, A. Weiss, J. Rheol.
45, 709 (2001).
22. P. D´ıaz-Leyva, M. Reufer, I. Lynch, J.L. Harden, F. Schef-fold, submitted.
23. J.J. Crassous, M. Siebenburger, M. Ballauff, M. Drechsler, D. Hajnal, O. Henrich, M. Fuchs, J. Chem. Phys. 128, 204902 (2008).
24. P.J. Flory, Principles of Polymer Chemistry (Cornell Uni-versity Press, New York, 1953).
25. I. Varga, T. Gilanyi, R. Meszaros, G. Filipcsei, M. Zrinyi, J. Phys. Chem. B 105, 9071 (2001).
26. M. Stieger, W. Richtering, J.A. Pedersen, P. Lindner, J. Chem. Phys. 120, 6197 (2004).
27. T. Hellweg, C.D. Dewhurst, W. Eimer, K. Kratz, Langmuir 20, 4330 (2004).
28. S. Seelenmeyer, I. Deike, S. Rosenfeldt, C. Norhausen, N. Dingenouts, M. Ballauff, T. Narayanan, P. Lindner, J. Chem. Phys. 114, 10471 (2001).
29. A.K. Lele, M.M. Hirve, M.V. Badinger, R.A. Mashelkar, Macromolecules 30, 157 (1997).
30. H.G. Schild, Progr. Polym. Sci. 17, 163 (1992).
31. W. McPhee, K.C. Tam, R. Pelton, J. Colloid Interface Sci. 156, 24 (1993).
32. A. Erbe, K. Tauer, R. Sigel, Phys. Rev. E 73, 031406 (2006).
33. R.S. Chen, H. Yang, X.H. Yan, Z.L. Wang, L. Li, Chem. J. Chin. U. 22, 1262 (2001).
34. J. Brandrup, E.H. Immergut, E.A. Grulke, A. Abe, D.R. Bloch (Editors), Polymer Handbook, 4th edition (John Wi-ley & Sons, 2005).
35. J.C. Maxwell-Garnett, Philos. Trans. R. Soc. London, Ser. A 203, 385 (1904).
36. M. Born, E. Wolf, Principles of Optics, 3rd edition (Perg-amon, Oxford, 1965).
37. B.J. Berne, R. Pecora, Dynamic Light Scattering (Wiley, New York, 1976).
38. P. Schurtenberger, M.E. Newman, in Environmental Par-ticles, edited by J. Buffle, H.P. van Leeuwen (Lewis Pub-lishers, Boca Raton, 1993), Chapt. 2, pp. 37-115.
39. C. Urban, P. Schurtenberger, J. Colloid Interface Sci. 207, 150 (1998).
40. H.C. van de Hulst, Light Scattering by Small Particles (Dover, New York, 1981); C.F. Bohren, D.R. Huffman, Ab-sorption and Scattering of Light by Small Particles (John Wiley & Sons, Inc., New York, 1983).
41. Polydisperse Mie data calculated using the MiePlot soft-ware (http://www.philiplaven.com/mieplot.htm). 42. A. Nieves, F.J. de las Nieves, A.
Fernandez-Barbero, J. Chem. Phys. 120, 374 (2004).
43. T.G. Mason, M.Y. Lin, Phys. Rev. E 71, 040801 (2005). 44. A. Ishimaru, Wave Propagation and Scattering in Random
Media (Academic Press, New York, 1978).
45. D.A. Weitz, D.J. Pine, Dynamic Light Scattering, edited by W. Brown (Oxford University Press, New York, 1993),
Chapt. 16, pp. 652-720.
46. S. Fraden, G. Maret, Phys. Rev. Lett. 65, 512 (1990). 47. P.D. Kaplan, M.H. Kao, A.G. Yodh, D.J. Pine, Appl. Opt.
32, 3828 (1993).
48. B.R. Dasgupta, S.Y. Tee, J.C. Crocker, B.J. Frisken, D.A. Weitz, Phys. Rev. E 65, 05505 (2002).
49. L.F. Rojas-Ochoa, S. Romer, F. Scheffold, P. Schurten-berger, Phys. Rev. E 65, 051403 (2002).
50. P.D. Kaplan, A.G. Yodh, D.J. Pine, Phys. Rev. Lett. 68, 393 (1992).
51. J. G´omez Rivas, D.H. Dau, A. Imhof, R. Sprik, B.P.J. Bret, P.M. Johnson, T.W. Hijmans, A. Lagendijk, Opt. Commun. 220, 17 (2003).
52. M. Reufer, L.F. Rojas-Ochoa, S. Eiden, J.J. S´aenz, F. Scheffold, Appl. Phys. Lett. 91, 171904 (2007).
